RESEARCH PAPER

Anti-coxsackievirus B4 (CV-B4) enhancing activity of serum associated withincreased viral load and pathology in mice reinfected with CV-B4

Firas Elmastoura,b,c, Hela Ja€ıdaneb,c, Mehdi Benkahlaa, Leila Aguech-Oueslatia,b, Famara Sanea, Aymen Halouanib,c,Ilka Engelmanna, Antoine Bertina, Moncef Moknid, Jawhar Gharbib, Mahjoub Aounib, Enagnon K. Alidjinoua,and Didier Hobera

aUniversit�e de Lille, Facult�e de M�edecine, CHU Lille, Laboratoire de Virologie/EA3610, Lille, France; bUniversit�e de Monastir, Laboratoire desMaladies Transmissibles et Substances Biologiquement Actives LR99ES27, Facult�e de Pharmacie de Monastir, Monastir, Tunisia; cUniversit�e deTunis El Manar, Facult�e des Sciences de Tunis, Tunis, Tunisia; dUniversit�e de Sousse, CHU Farhat Hached, Service d’Anatomopathologie, Sousse,Tunisia

ARTICLE HISTORYReceived 19 July 2016Revised 29 September 2016Accepted 18 October 2016

ABSTRACTIn previous studies it was shown that inoculation of Swiss albino mice with CV-B4 E2 resulted in theproduction of serum IgG capable of enhancing the CV-B4 E2 infection of murine spleen cellscultures. To investigate whether such an enhancing activity of serum can play a role in vivo, wedecided to study the CV-B4 E2 infection in mice exposed to successive inoculations of virus. In Swissalbino mice infected with CV-B4 E2 at the age of 21 days, anti-CV-B4 E2 neutralizing and enhancingactivities of their serum peaked after 55 d. In contrast, mice inoculated at the age of 55 d expressedmuch lower activities. Despite the neutralizing activity of serum, CV-B4 E2 inoculated a second timeto 55 day-old animals spread into the host. At the age of 72 and 89 d the levels of viral RNA andinfectious particles were higher in organs of animals exposed to 2 successive infections comparedwith animals infected once at the age of 21 d or 55 d. In animals with 2 successive inoculations ofCV-B4 E2 there was a relationship between the anti-CV-B4 E2 enhancing activity of serum and thelevel of viral RNA in organs and an enhancement of pathology was observed as displayed byhistological analysis of pancreas and hyperglycaemia. Altogether our data strongly suggest that ananti-CV-B4 E2 enhancing activity in the host can play a role in the outcome of a secondary infectionwith this virus.

KEYWORDSenterovirus; hyperglycaemia;in vivo; pancreas; reinfection

Introduction

The Enterovirus genus encompasses many human sero-types distributed in 7 species Enterovirus A to D and Rhi-novirus A to C.1 They are small, non-enveloped, positivesingle-stranded RNA viruses belonging to the Picorna-viridae family. Type B Coxsackievirus (CV-B) encom-passing CV-B 1–6, may be responsible for a largepathological specter varying from benign acute to severechronic diseases such as chronic myocarditis, dilated car-diomyopathy, and type 1 diabetes.2 An associationbetween enteroviruses, especially CV-B and type 1 diabe-tes (T1D) was reported. Enteroviral components havebeen more frequently detected in blood of T1D patients,and a meta-analysis has reported a strong associationbetween EV detection, especially CV-B, and T1D.3 Thepresence of EV protein in the pancreas of T1D patientshas been observed.4 Prospective studies confirmed thelink between EV and the development of pancreas islet

autoimmunity.5 Moreover, experimental studies, in vitroand in vivo animal models, revealed that CV-B4 may beinvolved in the pathogenesis of T1D through severalmechanisms.2,6,7,8

It has been shown that non-neutralizing anti-CV-B4IgG obtained from serum of patients can increase theinfection of peripheral blood mononuclear cells (PBMC)with CV-B4, which results in production of IFNa andother inflammatory cytokines.9,10,11,12,13,14 The role ofenhancing IgG levels in the outcome of CVB4 infectionsand other conditions such type 1 diabetes, cannot beignored.12,14 However the impact of anti-CV-B4 enhanc-ing activity of serum/IgG in the infection with this virus,in vivo, remains to be investigated.

A variant of CV-B4, called CV-B4 E2, isolated from apatient who died after onset of type 1 diabetes, was ableto induce diabetes in mice.15 It was reported that thisdiabetogenic strain CV-B4 E2 could infect outbred Swiss

CONTACT Didier Hober didier.hober@chru-lille.fr Laboratoire de Virologie EA3610, Center P Boulanger, Hopital A Calmette CHRU, 59037 Lille Cedex,France.© 2017 Taylor & Francis

VIRULENCE2017, VOL. 8, NO. 6, 908–923https://doi.org/10.1080/21505594.2016.1252018

albino mice.6 In Swiss albino mice infected with CV-B4E2 a high extent of anti-CV-B4 E2 neutralizing activitycan be observed, which is followed by a drastic decreaseto barely detectible levels.16 It was also shown that inocu-lation of mice with CV-B4 E2 resulted in the appearanceof IgG able to enhance the infection with this virus invitro. Indeed serum samples from Swiss albino miceinoculated with CV-B4 E2 intraperitoneally had anenhancing activity as demonstrated by assays usingmurine spleen cells cultures inoculated with CV-B4 E2mixed with various dilutions of serum. These assays indi-cate that serum from immune mice mixed with CV-B4E2 increases the antiviral titer, and the content of infec-tious particles in supernatants of spleen cells cultures, aswell as the level of intracellular viral RNA evaluated byRT-PCR. The enhancing activity was due to the IgG-enriched fraction of serum from CV-B4 E2 infected ani-mals, which was not observed with control animals asproved by isolating IgG from serum by using Protein Gaffinity chromatography.16

This observation deserves further investigation todetermine whether such an enhancing activity of serumcan play a role in vivo. Therefore we decided to study theCV-B4 E2 infection in mice exposed to successive inocu-lations of virus. The purpose was to determine in animalsreinfected with CV-B4 E2 the pattern of enhancing andneutralizing activity of serum and the relationship withthe viral load in organs. Moreover it has been investi-gated whether a previous infection can enhance thepathology induced by a second infection.

Results

Neutralizing and enhancing activities of serum fromSwiss albino mice inoculated with CV-B4 E2at the age of 21 or 55 d

Mice were inoculated with CV-B4 E2 (2 £ 104 TCID50)either at the age of 21 or 55 day-old. Serum was collectedfrom mice at the age of 21 or 55 d up to the age of 89 d(21, 28, 35, 42, 49, 55, 62, 72, 79 and 89 day-old). Ananti-CV-B4 neutralizing activity was found in serumsamples from each mouse. The titer was at least 1048when the animals were 55 day-old, and it was 64 whenthey were 89 day-old (Fig. 1a). When 55 day-old animalswere inoculated with CV-B4 E2 the resulting anti-CV-B4neutralizing activity of their serum was lower than 64(Fig. 1a).

The enhancing activity of serum samples from thesemice inoculated when they were 21 or 55 day-old wasdetermined. Virus suspensions (CV-B4 E2 at 105

TCID50/mL) were incubated in presence of mouse serumdiluted at 1/1000, then the mixtures were added to

murine spleen cells cultures. Supernatants were har-vested 48 hours later and checked for their antiviralactivity by using the bioassay described in the materialand methods section. In addition the level of intracellularenteroviral RNA in spleen cells harvested at 48 hours p.i.was evaluated by quantitative real-time RT-PCR. Basedon this, it was observed that serum samples from miceinoculated when they were 21 day-old enhanced the anti-viral activity of spleen cells up to at least a titer value of16 at the age of 55 day-old and afterwards the titer valuesdecreased. In contrast in serum from mice inoculatedwhen they were 55 day-old the values were lower than 2(Fig. 1b).

Moreover the serum of mice infected when they were21 day-old could enhance the CV-B4 E2 infection ofspleen cells cultures as demonstrated by the detection ofintracellular viral RNA by RT-PCR. The level of intracel-lular viral RNA was at least 3.5 £ 105copies / ng of totalRNA when serum samples were obtained from thesemice at the age of 55 day-old afterwards the values weredecreasing. In serum from mice inoculated when theywere 55 day-old the resulting enhancing activity wasweak, indeed with serum samples from these animals thelevel of intracellular viral RNA was lower than 35 copies/ng of total RNA which was similar to the values obtainedwhen spleen cells cultures were exposed to CV-B4 E2 (30copies/ng of total RNA) (Fig. 1c).

Thus in each mouse inoculated with CV-B4 E2 at theage of 21 day-old, the levels of neutralizing and enhanc-ing activities of serum when they were 55 day-old werehigh. In contrast in each mouse inoculated with CV-B4E2 at the age of 55 day-old, the levels of neutralizing andenhancing activities of serum were low. Therefore it wasdecided to investigate the pattern of infection in miceinoculated when they were 21 day-old and inoculatedagain when they were 55 day-old.

Viral load in mice inoculated with CV-B4 E2 whenthey were at the age of 21 d or / and 55 d

Mice were gathered in 3 groups encompassing 25 ani-mals in each. Mice were inoculated intraperitoneallywith CV-B4 E2 (2 £ 104 TCID50) either once when theywere 21 day-old (group 1), or when they were 55 day-old(group 2), or twice, when they were 21 and then 55 day-old (2 successive infections) (group 3). In each animalsmall volumes of blood samples were collected in orderto investigate the anti-CV-B4 E2 neutralizing andenhancing activities of serum. The enhancing activitywas determined by a bioassay through the level of CV-B4 E2 RNA in spleen cell in vitro as detailed above. Fivesanimals were sacrificed at the age of 21, 38, 55, 72 and89 day-old and from each mouse 7 organs were collected.

VIRULENCE 909

Samples of the 7 organs were treated to extract ARN andto evaluate the level of enteroviral RNA by quantitativereal-time RT-PCR and to isolate infectious particles. Thedetection of infectious particles and of viral RNA inserum at the age of 38, 55, 72 and 89 d was negative inanimals inoculated with the virus at the age of 21 d or55 d or both (data not shown).

In mice inoculated at the age of 21 day-old (group 1),the anti-CVB4 E2 neutralizing and enhancing activities inserum peaked when they were 55 day-old as expected fol-lowing the results obtained in the first part of this study(Fig. 2a and 2b). Viral RNA was detected in the first sam-ple (collected at the age of 38 days) and in the next one (atthe age of 55 days) in each organ afterwards the levels

Figure 1. Continued

910 F. ELMASTOUR ET AL.

Figure 1. Individual follow-up representation of anti-CV-B4 E2 neutralizing and enhancing activity of serum from CV-B4 E2-inoculatedmice. Blood samples were successively collected from 10 mice inoculated with 2£104 TCID50 CV-B4 E2 at the age of 21 d (cross), and10 mice inoculated at the age of 55 d (square). Serum was recovered and each serum sample was tested for neutralizing activity (a) andenhancing activity (b and c). a) Neutralizing activity. Neutralizing titers of serum samples were defined as the reciprocal of the last dilu-tion that totally inhibited the CVB4 E2-induced CPE in HEp-2 cell cultures. b) Enhancing activity tested through the antiviral activity insupernatants of spleen cells cultures inoculated with CV-B4 E2 mixed with serum from CV-B4 E2-infected mice. Serum samples werediluted at a ratio of 1: 500 and mixed with CV-B4 E2. The mixture was then inoculated to murine spleen cells cultures (MOI D 0.02) andsupernatants were harvested after 48 hours of incubation. The antiviral activity of supernatants was tested using a bioassay. The antiviraltiters of supernatant samples were defined as the reciprocal of the last dilution that totally inhibited the EMCV-induced cytopathic effectin L-929 cell cultures. c) Enhancing activity tested through the quantification of intracellular viral RNA in spleen cells cultures inoculatedwith CV-B4 E2 mixed with serum (diluted 1:1,000) from CV-B4 E2-infected mice. Total RNA was extracted and the level of viral RNA wasdetermined by real time QPCR. The results were expressed as viral RNA copy number per ng of total RNA.

VIRULENCE 911

decreased dramatically or were null. The maximum val-ues were lower than 102 in most organs and lower than103 in the heart. Infectious particles were isolated fromevery organ in agreement with the detection of viral RNA.The levels of infectious particles expressed as PFU per mgpeaked in organ samples collected when mice were

55 day-old (or 72 day-old in the case of pancreas), the val-ues ranged from 102 to 104, which decreased to less than101 in brain, spinal cord and thymus samples obtainedfrom mice at the age of 89 day-old whereas in otherorgans (pancreas, intestine, heart and spleen) at this agethe values were around 103 (Fig. 2c).

Figure 2. Continued

912 F. ELMASTOUR ET AL.

In mice inoculated with CV-B4 E2 at the age of55 day-old (group 2), the anti-CVB4 neutralizing andenhancing activities in serum were very low (< 32 and <

2 respectively) which is in agreement with the resultsobtained in the first part of this study (Fig. 2a and b).Viral RNA was detected in the first sample (collected atthe age of 72 days) and in the next one (at the age of89 days) in pancreas and spleen, the values were < 102.Viral RNA was also detected in heart and thymus, thevalues were < 101, but was not detected in brain, spinalcord and intestine. The levels of infectious particles werelow < 102 in spleen, pancreas and heart obtained fromthose animals at the age of 72 and 89 day-old, and werelower in other organs in agreement with the viral RNAvalues (Fig. 2c’).

In mice inoculated at the age of 21 d with CV-B4 E2and inoculated a second time at the age of 55 d (group3), the anti-CV-B4 E2 neutralizing activity peaked at theage of 72 d (1638 C/¡ 560 and then decreased at the ageof 89 d (921 C/¡163) (Fig. 2 a”). At these ages theenhancing activity of serum was high as shown by thelevel of viral RNA in spleen cells cultures inoculated withmixtures of CV-B4 E2 and serum samples. The valueswere much higher than those obtained with serum frommice at the same ages that were infected once at 21-dayold (2.5 107 C/¡ 900 vs 6.103 C/¡ 1000 at the age of72 d p < 0.05; 1.4. 107 C/¡ 800 vs 6.102 C/¡ 160 at theage of 89 d p < 0.05). Viral RNA was detected in eachorgan, and the levels at the age of 72 d and 89 d in miceinoculated twice with CV-B4 E2 (group 3) were higher

Figure 2. Biological activity of serum from mice inoculated once or twice with CV-B4 E2 and viral parameters in their organs. FemaleSwiss Albino mice were inoculated with CV-B4 E2 at the age of 21 d (group 1), 55 d (group 2), or inoculated successively at the age of21 d and 55 d (group 3). Each inoculum was 2 10£4 TCID50. Group 1 and group 2 encompassed 25 animals each, while group 3 encom-passed 15 animals. Five animals were sacrificed at the ages of 21, 38, 55, 72 and 89 d (groups 1 and 3) and at the ages of 55, 72 and 89 d(groups 2). a – a”) Neutralizing activity of serum samples (see legend of figure 1). The results are mean C SD, n D 5. b – b”) Enhancingactivity tested through the quantification of intracellular viral RNA in spleen cells cultures inoculated with CV-B4 E2 mixed with serum(see legend of figure 1). The results are mean C SD, n D 5. c – c”) Viral parameters in organs. A sample of each organ collected in miceinoculated with CV-B4 E2 was processed as described in the materials and methods section to determine the level of viral RNAexpressed as copy number per ng of total RNA (black bars) and the level of infectious particles expressed as pfu/mg (white bars). Theresults are mean C SD, n D 5. Arrow: CV-B4 E2 inoculation. NS: not significant (p value type D “Other” > 0.05). The data obtained atthe age of 72 and 89 d in mice inoculated once with the virus at the age of 21 d or inoculated twice with the virus at the age of 21 dand 55 d were analyzed. In most cases values obtained in mice inoculated twice were significantly higher than those obtained in miceinoculated once. Only the not significant (NS) differences are indicated.

VIRULENCE 913

Figure 3. (For figure legend, see page 915.)

914 F. ELMASTOUR ET AL.

than those obtained at the same age in heart, pancreas,spleen and intestine of mice inoculated once at the age of21 d (group 1) p values ranging from 0.022 to < 0.001.As far as the other organs the pattern of results at theages of 72 d and 89 d were not so clear: p values were0.057 and 0.001 respectively in the case of thymus, <0.01 and 0.06 in the case of brain, and 0.4 and 0.045 inthe case of spinal cord. Nevertheless, the viral RNA levelsin all organs were much higher than those obtained atthe same age in organs of mice inoculated once at theage of 55 d (group 2) (p values < 0.05 (Fig. 2c”).

The levels of infectious particles in heart, pancreas,intestine, spleen and thymus obtained at the ages of 72and 89 d from mice inoculated twice were higher than inthose organs obtained at the same ages from mice inocu-lated once (p value < 0.01). In contrast there was no sig-nificant difference in the levels of infectious particles inthe brain and spinal cord in animals which were inocu-lated once or twice with the virus (see Fig. 2).

Relationship between the anti-CV-B4 E2 enhancingactivity of serum and the level of viral RNA in organs

Viral RNA was detected in organs of mice inoculated atthe age of 21 d with lower levels than mice inoculatedsuccessively at the age of 21 and 55 days, whereas theywere very low or null in organs of mice inoculated at theage of 55 d. In mice inoculated later in life the enhancingactivity of their serum was very low or null comparedwith animals inoculated at the age of 21 d or inoculatedtwice. These observations prompted us to study the rela-tionship between the serum enhancing activity and thelevel of viral RNA in organs.

The enhancing activity of serum measured throughthe viral RNA copy number in spleen cell cultures, inoc-ulated with CV-B4 E2 mixed with serum (diluted 1/1000), was compared with the level of viral RNA inorgan samples in animals infected at the age of 21 d.A significant correlation, (p value < 0.0001), betweenserum activity and intestinal viral load was only obtainedwhen the animals were 72 day-old (Fig. 3a). Howeverthere was no correlation between serum activity andintestinal viral load or other organs (brain, spinal cord,spleen and thymus, heart and pancreas) when the ani-mals were 38, 55, 72 or 89 day-old (data not shown).

When the parameters were compared in animalsinfected successively at the age of 21 and 55 days, a sig-nificant correlation was obtained between serum activityand viral load in spleen (p value < 0.0001) and intestine(p value < 0.0001), collected when the animals were72 day-old, and viral load in heart and pancreas collectedwhen the animals were 72 day-old and 89 day-old as well(p value < 0.003) (Fig. 3b). However no correlation wasobserved between serum activity and viral load in brain,spinal cord and thymus when collected at the same age(data not shown).

A positive correlation was obtained between the anti-CV-B4 E2 enhancing activity of serum and the viral loadin organs at the age of 72 d and/or 89 d, whereas therewas no relationship between the anti-CV-B4 E2 neutral-izing activity of serum and the level of viral RNA inorgans (data not shown).

The relationship between the anti-CV-B4 E2 serumactivity, enhancing and neutralizing, and the level ofinfectious particles isolated from organ samples wasstudied. There was no significant positive correlationbetween these parameters in animals inoculated once atage of 21 d or inoculated successively at the age of 21and 55 d (data not shown).

Enhanced pathology in mice inoculated with CV-B4E2 at 21 d and 55 d of age successively

There was no evidence of heart, pancreas (Fig. 4a) andbrain damage in control mice (n D 5) and in mice inocu-lated once with CV-B4 E2 at the age of 55 d (n D 5).There was no lesion of organs in mice inoculated at theage of 21 d when they were 38 (n D 5) and 55 day-old(n D 5) (data not shown). Little or no inflammation wasobserved in heart from one out of 5 mice at the age of72 d and in pancreas from 2 out of 5 animals at this age(Fig. 4b) and from one out of 5 at the age of 89 d (datanot shown).

In marked contrast loss of exocrine tissue replacementwith fat tissue and infiltration of mononuclear cells wereobserved in pancreas from 5 out of 5 mice at the age of72 d (Fig. 4c) and from 3 out of 5 mice at the age of 89 d(data not shown) when the animals were inoculated suc-cessively at the age of 21 and 55 d. An extensive

Figure 3. (see previous page) Relationship between viral load and anti-CV-B4 E2 enhancing activity of serum in mice. Female SwissAlbino mice were inoculated with CV-B4 E2 at the age of 21 d and serum and organs were collected when they were at the age of 72 d(n D 5) (a), or mice were inoculated successively when they were at the age of 21 and 55 d and serum and organs were collected from5 animals when they were at the age of 72 d and from 5 animals when they were at the age of 89 d (b). Each inoculum was 2 10 £ 4TCID50. For each mouse both anti-CV-B4 E2 enhancing activity of serum and viral load in organs were measured. The anti-CV-B4 E2enhancing activity of serum was tested through the quantification of intracellular viral RNA in spleen cells cultures inoculated with CV-B4 E2 mixed with serum (see legend of figure 1). Total RNA was extracted from spleen cells and the level of viral RNA was determinedby real time QPCR. The results were expressed as viral RNA copy number per ng of total RNA.

VIRULENCE 915

infiltration of mononuclear cells was seen in the heartcollected at the age of 72 d from one out of 5 mice inocu-lated twice with CV-B4 E2 (Fig. 4c).

In contrast, there were no apparent lesions in thebrain at the age of 72 or 89 d in any animal inoculatedonce or twice with CV-B4 E2 (data not shown).

To investigate further the impact of CV-B4 E2 ontothe host, we measured the serum levels of glucose in ani-mals as described in the materials and methods section.In animals infected twice, the serum levels of glucose atthe age of 72 and 89 d were 439 C/¡ 80 mg/dl (n D 5)and 393 C/¡ 56.72 mg/dl (n D 5) respectively and were

Figure 4. Histologic examination of organs from mice. Female Swiss Albino mice were inoculated with culture medium (a), or CV-B4 E2when they were 21 d old (b), or inoculated successively when they were 21 and 55 d old (c). Each inoculum was 2 10 £ 4 TCID50. Ani-mals were sacrificed and organs were collected and processed as described in the material and methods section. The slices weremounted and observed under Zeiss Axioscan(Z1) Scanner. The pictures represent pancreas and heart obtained from animals at the ageof 72 d. The arrows show inflammatory foci.

916 F. ELMASTOUR ET AL.

significantly higher than the values measured at the sameage in serum of animals infected only once at the age of21 d (see Fig. 5). Indeed the mean levels of glucose inserum from 21 through 89 d of age in these animals werelower than 200 C/¡ SD mg/dl. The pattern of results

was similar in serum from mice inoculated when theywere 55 day-old and in control animals from 21 through89 d of age with mean values ranging from 162 C/¡ 54to 221 C/¡ 39 mg/dl (see Fig. 5).

Discussion

The current study addressed the issue of successive infec-tions with CV-B4 E2 in mice.

Several considerations are noteworthy. Recently ourteam reported that serum samples from mice inoculatedwith CV-B4 E2 enhanced the infection of spleen cellswith this virus.16 In our previous report the infection ofspleen cells inoculated with CVB4 mixed with dilutedimmune serum was assessed by the detection of viralRNA by RT-PCR, whereas viral RNA was only detectedby sn-RT-PCR, a more sensitive technique, when cellswhere inoculated with CV-B4 E2. In the present study,the levels of intracellular viral RNA in spleen cells weremeasured by quantitative real-time RT-PCR. The quanti-fication of viral RNA in this bioassay displayed theenhancing activity of diluted serum and enabled evalua-tion of the extent of this property in each serum sampleobtained successively up to the age of 89 d. Moreover thequantitative evaluation of the enhancing activity ofserum samples allowed inter-individual comparisonsand allowed the investigation of the relationship betweenserum enhancing activity and the level of viral RNA inorgan samples.

This study shows that mice infected at the age of 21and 55 d have higher levels of viral RNA and infectiousparticles at the age of 72 and 89 days, as proved by thehigher values of viral parameters in these mice whencompared with mice infected only once at the age of21 d. The level of viral markers in mice infected twice isnot due to the sum of the parameters observed at the ageof 21 and 55 days, the latter case exhibiting very lowvalues.

The level of viral RNA in our experiments wasexpressed as copy of viral RNA / ng of total RNA in tis-sue. Considering the amount of RNA in 1 mg of tissuethe level of viral RNA per mg of tissue was highly

Figure 5. Glucose levels in serum. Female Swiss Albino mice wereinoculated with CV-B4 E2 when they were at the age of 21 d (a),56 d (b), or inoculated successively when they were 21 and 56 d(c). Each inoculum was 2 10 £ 4 TCID50. Uninfected animalsserved as controls (d). In each case 5 animals were sacrificed atthe age indicated in the x axis. Blood was collected; serum wasobtained and stored at ¡80�C. Then the serum levels of glucosewere measured on a glucometer as described in the materialsand methods section. The results are means C SD (n D 5).�p value 0.001 ��p value 0.02 vs serum level of glucose in mice atthe age of 72 and 89 d infected once at the age of 21 d (a).

VIRULENCE 917

multiplied (> 100). Thus the amount of viral RNA com-pared with the level of PFU per mg of tissue is in agree-ment with the fact that picornaviruses produce an excessof RNA compared with infectious particles. There wasno correlation between the levels of viral RNA and infec-tious particles in organ samples in our experiment. Wehave no clear explanation; a discrepancy between theseparameters can be due to a poor recovery rate of infec-tious particles and/or viral RNA. Moreover, it cannot beruled out that neutralizing specific antibodies in the sam-ples, at high levels in certain individuals, as reflected bythe results of sero-neutralization assays, impaired theisolation of infectious particles, even though there wasno inverse correlation between the anti-CV-B4 E2 neu-tralizing titres and the amounts of isolated infectiousparticles (data not shown).

In mice infected at a late stage, at the age of 55 days,the levels of viral RNA and infectious particles in organswere very low compared with mice infected at an earlierstage at the age of 21 d. The moderate infection in oldermice was not due to a stronger immune response result-ing in a more efficient viral clearance since the level ofanti-CV-B4 E2 neutralizing activity in their serum waslower than in other mice infected at the age of 21 d. Thispattern of CV-B4 E2 infection in older mice can be dueto a natural resistance to the virus at late stages of lifepossibly because the levels of the virus receptor, the cox-sackievirus and adenovirus receptor (CAR) falls signifi-cantly with age.17 Interestingly, however, in older miceinfected with CV-B4 E2, it has been observed that theanti-CVB4 enhancing activity of their serum was verylow. Whether this pattern of enhancing activity plays arole in the low viral load in older animals remains anopen question, but this observation suggests that mostprobably successive viral infections in older mice wouldnot result in high viral loads in organs and consequentlyyoung individuals, with a higher enhancing activity oftheir serum, are more prone to an increased viral loadfollowing repeated infections. Furthermore as CAR isknown to be upregulated in stressed tissues,18 anincreased expression of CAR following the first infectionmay to some extent participate in the enhanced viraltiters in these tissues observed in reinfected mice.

It is intriguing that 21 day-old mice inoculated withCV-B4 E2 were successfully infected a second time atthe age of 55 days, as proved by the high viral load inorgans in the following days, even though the anti-CV-B4 E2 neutralizing activity of serum was strong whenthe virus was administered. This pre-existing neutraliz-ing activity boosted by the second virus inoculation didnot prevent the production of the virus as proved bythe isolation of viral particles from organs in ourexperiments. Altogether, our data suggest that the anti-

CV-B4 E2 enhancing activity in an individual can playa role in the outcome of a secondary infection with CV-B4 E2 in the host. Indeed, as demonstrated in thisstudy, serum samples from 55 day-old mice that wereinfected at age of 21 d were able to enhance dramati-cally the infection of cells with CVB4 in vitro. More-over, a relationship between the anti-CV-B4 E2enhancing activity of serum, and the levels of viralRNA in organs collected at the age of 72 and 89 d frommice infected twice was observed. Thus it cannot beexcluded that under favorable condition, reflected bythe anti-CV-B4 E2 enhancing activity of serum, a sec-ondary exposure to CV B4 E2 result in a successfulinfection with a high viral load in organs despite astrong neutralizing activity of serum.

It can be assumed that the anti-CV-B4 E2 enhancingactivity of serum displayed in vitro was due to anti-CV-B4 E2 IgG as reported in our previous studies.16 Thus,such IgG were present in serum samples from miceinoculated with CV-B4 E2 but was not detected in theirserum before exposure to the virus as shown in thepresent study when the individual anti-CV-B4 E2enhancing activity was tested in consecutive serumsamples from 10 animals. These observations stronglysuggest that the anti-CV-B4 E2 enhancing activity ofserum was due to anti-CV-B4 E2 antibodies, and thatthese antibodies played a role in the pattern of infectionof animals in vivo. Further studies are needed to inves-tigate whether anti-CV-B4 E2 antibodies are involvedin the increased level of viral markers in animals inocu-lated with the virus, and particularly in mice infectedwith CV-B4 E2 successively at the age of 21 d and 55 din our experiments.

Histological examination of tissue from mice in ourexperiments showed lesions of pancreas, and of heart butat a lower extent, that were more prevalent and/or pro-nounced in samples obtained from 72-day-old miceinfected successively at the age of 21 and 55 d comparedwith samples obtained from mice at the same age, or atthe age of 38 and 55 days, infected only once at the ageof 21 d.

This impact of CV-B4 E2 onto the tissues in miceinoculated twice was surprising. Indeed, it was expectedthat mice with high anti-CV-B4 E2 neutralizing activityin serum were protected against lesions of organs whenthe virus was inoculated a second time. It cannot beexcluded that the neutralizing activity in this case limitedthe spreading and/or the impact of the virus. Neverthe-less the animals were not protected against lesions oforgans in these conditions. The extent of the lesions washigh in pancreas of mice inoculated twice with CV-B4E2; whereas in brain there were no apparent injuries. Inthe pancreas, but not in the brain of mice inoculated

918 F. ELMASTOUR ET AL.

twice, the viral load was high and related to the enhanc-ing activity of serum. Altogether this data suggests thatlesions of the pancreas were related to the viral load intissues, the extent of which depended on enhancingantibodies.

The serum level of glucose measured on a glucom-eter was high in mice at 72 and 89 d of age, infectedtwice with CV-B4 E2 (> 300 mg/dl) whereas therewas no difference between serum levels of glucose inmice infected only once and controls groups ( around200 mg/dl). The glycaemia in mice was already mea-sured on a standard glucometer19,20 and animals wereconsidered hyperglycemic when readings were greaterthan 200 mg/dl in previous studies. In these studiesthe glucose levels were measured using bloodobtained from the tail vein or eye bleeds, in contrastour study used serum samples obtained from bloodof sacrificed animals. There is a discrepancy betweenour results and those of other authors regarding theimpact of CV-B4 E2 onto the glucose homeostasis.20

In our experiments, hyperglycemia was expressed infemale Swiss albino mice exposed to CV-B4 E2 twice,but there was no disturbance from 17 d through upto 70 d post-inoculation in these mice challengedonce intraperitoneally with 2 £ 104 TCID50 at theage of 21 day-old. In contrast, in the previous studyhyperglycemia was observed starting from 49 d afterinfection of male Swiss-derived CD1 mice with 10 £4 PFU CV-B4 E2 when they were 28–35 day-old.20 Inthe previous study CV-B4 E2 stocks were prepared inmonolayers of monkey kidney cells whereas in ourexperiments virus stocks were prepared in humanHEp-2 cell cultures. Whether such differences, in con-junction with the type and gender of mice, canexplain the discrepancy remains to be determined.

It was reported that in 6 week-old SJL mice, a homol-ogous challenge intraperitoneally with 10 £ 4 pfu CV-B4E2 6 weeks after the first inoculation resulted in anenhancement of hyperglycaemia independently of the Tcell response.19 This could be due to an antibody-mediated enhancement of infectivity though antibodiesable to increase the virus spreading and replicationresulting in hyperglycaemia as suggested by the relation-ship between the anti-CV-B4 E2 enhancing activity ofserum, the levels of viral RNA in organs, especially thepancreas, and the disturbance of glycaemia homeostasisobserved in mice inoculated twice in the current study.

Based on the results of our study, it cannot be ruledout that an anti-CV-B enhancing activity in the hostplayed a role in enhanced pathology observed inheterologous CV-B challenge of mice as previouslyreported.22,23,24,25 Further studies are needed to addressthis issue.

The relationship between the anti-CV-B4 E2 enhanc-ing activity of serum and the levels of viral RNA inorgans, combined with the observation of enhancedpathology as displayed by tissue lesions and hyperglycae-mia in mice challenged with CV-B4 E2 in the currentstudy, suggest that an antibody-mediated enhancementof CV-B4 E2 infectivity could increase the virus spread-ing and replication resulting in a more severe outcomeof the infection. Based on this hypothesis, the anti-CV-B4 enhancing activity in serum of human beings previ-ously described by our team could be considered as arisk factor for enhanced pathology in case of exposure toCV-B4.9,10,11 Similarly an enhancing activity toward anyCV-B would be a risk factor for enhanced pathology incase of homologous, and possibly, heterologous chal-lenge with CV-B. Interestingly, in patients with type 1diabetes, an anti-CV-B4 enhancing activity of serum wasobserved.11,26,27 Whether enhancing antibodies play arole in the pathogenesis of CV-B –induced diseases invivo, especially type 1diabetes deserves further studies.Future studies will be directed along this line in ourlaboratory.

Material & methods

Virus and cells

The diabetogenic strain CV-B4 E2 (kindly provided by J.W. Yoon, Julia McFarlane Diabetes Research Center,Calgary, Alberta, Canada), was propagated in HEp-2cells (BioWhittaker, Vervier, Belgium) in Eagle’s mini-mal essential medium (MEM) supplemented with 10%fetal calf serum, 1% L-glutamine (Gibco BRL),50 mg/mLstreptomycin, and 50 IU/mL penicillin (Invitrogen, SaintAubin, France). The encephalomyocarditis virus (EMCV)strain (ATCC) was propagated in L-929 cells (kindlyprovided by T Jouault, Lille, France) in Dulbecco’s modi-fied eagle’s medium (DMEM; Gibco BRL, Invitrogen)supplemented with 10% fetal calf serum, 1% L-gluta-mine, 50 mg/mL streptomycin, and 50 IU/mL penicillin.Supernatants were collected 3 d after inoculation, clari-fied by centrifugation at 2,000 £ g for 10 min dividedinto aliquots, and stored at ¡80�C.

Virus titers in stocks were determined on HEp-2 cells(CV-B4 E2) or L-929 cells (EMCV) by limiting dilutionassay for 50% tissue culture infectious doses (TCID50)and expressed as log10 TCID50/mL as describedelsewhere.15

Animals

All animals were handled by respecting the standards ofgeneral ethics guidelines, and maintained in specific

VIRULENCE 919

“pathogen-free” conditions with unlimited access to ster-ile food and water. All experiments were conducted fol-lowing the guidelines of 2010 EU directory and wereapproved by Faculty of Pharmacy University of Monastir(Tunisia) and by the Ethical Committee for AnimalExperimentation of Nord-Pas-de-Calais (France).

Three -week-old Female Swiss albino mice (InstitutPasteur, Tunis were intraperitoneally inoculated with200 mL culture medium or with CV-B4 E2 at 2 £ 104

TCID50 units contained in 200 mL culture medium. Micewere followed up to the age of 89 d. When appropriate,blood samples were obtained from the tail vein of ani-mals, using a needle. Serum samples obtained followingcentrifugation of blood were stored at ¡80�C.

Other mice followed up to the age of 89 d were sacri-ficed using isoflurane at the ages of 21, 55, 72 and 89 d tocollect blood and organs (heart, pancreas, intestine,spleen, thymus, brain and spinal cord). Serum samplesobtained from blood were stored at ¡80�C. A part of theorgan sample was stored at ¡80�C for investigating thelevel of viral RNA and infectious particles, and anotherpart was fixed in formalin for histological study.

Seroneutralization assay

A modified Seroneutralization assay was used to assessthe anti-CV-B4 E2 activity of mice sera. HEp-2 cellswere first seeded at 105 cells per well, in 96-well microti-ter plates (Falcon, Oxnard, CA, USA) which were placedat 37�C in a humidified incubator with 5% CO2. After16 h of incubation, the medium was discarded and50 mL of culture medium was added in each well. Then5 mL of MEM or mouse serum samples, mixed with45 ml of MEM (1:10 dilution to prevent cytotoxicity),was added to the first wells, and 2-fold serial dilutionswere performed. Each sample was tested in triplicate.Afterwards, 25 mL of culture medium containing 25TCID50 CV-B4 and 25 ml of medium culture wereadded to each well and the plates returned to the incuba-tor. The cell layers were examined 48 hr later under aninverted microscope (Olympus CKX41) at the objectivelens 10x. The results were expressed as the inverse of thefinal dilution that inhibited the viral cytopathic effect.

Spleen cells cultures

The CV-B4 E2 enhancing activity of serum from micewas tested in murine spleen cells cultures as previouslydescribed.16

Spleens were aseptically removed from 5-week-oldmale Swiss albino mice (Janvier Laboratories, France)sacrificed by cervical dislocation. Spleen cells were pre-pared on ice, depleted of erythrocytes by hypotonic

shock, then suspended in RPMI-1640 (Eurobio, Paris,France) supplemented with 10% FCS, 1% L-glutamine,50 mg/mL streptomycin, 50 IU/mL penicillin and 10¡5Mb-mercaptoethanol (Sigma). Cells were then seeded into96-well culture plates at 5 £ 105 cells per well, and incu-bated at 37�C in a humidified atmosphere with 5% CO2.

Mouse serum samples (5 mL) were diluted at 1:1000in duplicate in supplemented RPMI and mixed (2/1; vol/vol) with CV-B4 E2 suspension (105 TCID50/mL). Themixtures were incubated for 2h at 37�C, then 85 mL ofthese mixtures were inoculated in duplicate to spleencells cultures (5 £ 105 in 100 ml/ well) and 100 ml of sup-plemented medium was added before incubating themicroplates for 48 h. Afterwards, culture supernatantswere collected, clarified and stored at ¡80�C for antiviralactivity titration. Furthermore, spleen cells cultures werewashed 5 times with cold phosphate buffered saline(PBS) and then prepared for RNA extraction followed byreal-time QPCR to quantify the level of intracellular viralRNA.

Bioassay for evaluating the antiviral activity ofsupernatants of spleen cells cultures

The antiviral activity of supernatants of spleen cells cul-tures was evaluated using a bioassay based on the protec-tion of L929 cells against the cytopathic effect (CPE)induced by EMCV as previously described.14 The inverseof the highest dilution providing 100% protection of thecells from EMCV-induced CPE was considered as theendpoint for antiviral activity.

Detection of infectious particlesInfectious particles were isolated from organ samplesand the amount was evaluated as previously described.19

Briefly, snap frozen tissues were weighed and crushedusing a tissue ruptor (Qiagen) in 1 ml PBS (1% penicil-lin/streptomycin), and then centrifuged at 2000 g for10 min at 4�C. Supernatants were 10-fold diluted inMEM with 2% FCS and 500 ml and were inoculated toconfluent HEp-2 cell layers in 6-well plates (Falcon,Oxnard, CA, USA) and 500 ml was used for RNA extrac-tion. After 24 h of incubation at 37�C in a humidifiedincubator with 5% CO2, the medium was discarded and200 ml of MEM containing crushed organs samples(dilution 10¡1 to 10¡9) were added to each well, andthen the plates were incubated for 3 hours. Afterwardsthe supernatants were discarded and the cell layers werewashed 5 times with culture medium, and then overlaidwith culture medium containing agarose (1:10 dilution).The plates were kept at room temperature until the agaroverlay turned solid. The plates were incubated, after 3–5 days, 1 mL of formaldehyde was added and 2h latter at

920 F. ELMASTOUR ET AL.

room temperature, the agarose overlay was discardedbefore staining the cell layers with crystal violet. The pla-ques were counted and the results were expressed asmeans of plaque forming units (PFU)/mg of tissue.

Quantification of viral RNA

RNA extraction

RNA was extracted from mice sampled tissues asdescribed above and from 3 £ 106 spleen cells by theacid guanidium thiocyanate-phenol-chloroform extrac-tion procedure by using Tri-Reagent (Sigma), asdescribed elsewhere.16 Purified water for injection (C.O.M Lavoisier) was submitted to the same extraction pro-cedure and served as negative control. Supernatant ofCV-B4 E2-infected HEp-2 cells served as positive con-trol. Extracted RNA was dissolved in 50 ml of nuclease-free water (Promega), quantified using the Quant-iTRiboGreen RNA assay kit (Molecular Probes, Invitrogen)according to the manufacturer’s instructions, The assayis performed in a microplate read by a Mx3000PTM

( Stratag�ene�) and extracted RNA was diluted to be usedin reverse transcription (RT)-PCR assays.

Quantitative RT-PCR

The Affinityscript� QPCR cDNA Synthesis Kit (Agilent)was used for the ARN retrotranscription step on a PerkinElmer 2400 thermocycler following the cycle conditions,25�C for 5min, 42�C for 15min and 95�C for 5 min.Quantitative RT-PCR for cDNA amplification was per-formed with the Brilliant� II QPCR Kit (Agilent) as rec-ommended by the manufacturer, under common cycleconditions (10 min at 95�C, 40 cycles of 30s at 60�C onthe Mx3000p thermal cycler (Stratagene). The primerswere as follows: sens, 50-CCC TGA ATG CGG CTAATC-30 and reverse, 50-ATT GTC ACC ATA AGCAGC-30. The sequence of the probe was FAM-AACCGA CTA CTT TGG GTG TCC GTG TTT-TAMRA.Enterovirus 71 RNA (Vircell) was used as standard forquantification.10

Histological analysis

Tissues were fixed with Formol embedded with paraffinand then cut into 4 mm thick sections, The slices weredeparaffinized twice in xylene for 10 min, then rehy-drated through sequential concentrations of alcohol(100, 95, 70 and 50�) and finally with double distilledwater. Afterward, slices were stained with hematoxylinfor 15 min, washed in running water, differentiated with1% HCL/ethanol and incubated in water for 5 min. For

cytoplasm staining, slices were then immersed in 0.5 %eosin solution for 3 min and washed in water for 10 min.Sections were dehydrated again through sequential con-centrations of alcohol (50, 70, 95 and 100�). Finally,slides were mounted and observed under Zeiss Axioscan( Z1) Scanner.

Serum level of glucose

Glucose level in serum of mice was measured with con-tour � XT from BAYER according to the manufacturer’sinstructions. Two ml of serum were put on strips so that0.6 ml was aspirated by capillarity. The results wereexpressed as mg/dl.

Statistical analysis

Graphs and analyses, Spearman test and t test whenappropriate, were performed with GraphPad Prism V6.0software, significance set at 0.05.

Disclosure of potential conflicts of interest

No potential conflicts of interest were disclosed.

Acknowledgments

The authors thank Meryem Tardivel of the BICeL-CampusHU Facility of Lille for access to systems and technical advice.The authors thank D. Chetcuti for reading the manuscript.

Funding

This study was supported by : Minist�ere de l’Education Natio-nale de la Recherche et de la Technologie, Universit�e Lille 2(Equipe d’Accueil 3610), France, and Center HospitalierR�egional et Universitaire de Lille, and EU FP7 (GA-261441-PEVNET: Persistent virus infection as a cause of pathogenicinflammation in type 1 diabetes - an innovative research pro-gram of biobanks and expertise). The Comit�e Mixte de Coop�e-ration Universitaire Franco-Tunisien (CMCU 08/G808) andEgide Paris. Minist�ere de l’Enseignement Sup�erieur et de laRecherche Scientifique, (LR99ES27), Tunisia.

References

[1] Knowles NJ, Hovi T, Hyypi€a T, King AMQ, Lindberg M,Pallansch MA, Palmenberg AC, Simmonds P, Skern T,Stanway G, et al. 2012. Picornaviridae. In Virus Taxon-omy: Classification and Nomenclature of Viruses: NinthReport of the International Committee on Taxonomy ofViruses; King AMQ, Adams MJ, Carstens EB, LefkowitzEJ, Eds; Publisher: San Diego, CA, 2012; pp. 855-880

[2] Ja€ıdane H, Sauter P, Sane F, Goffard A, Gharbi J, HoberD. Enteroviruses and type 1 diabetes: towards a better

VIRULENCE 921

understanding of the relationship. Rev Med Virol 2010;20:265-80; https://doi.org/10.1002/rmv.647

[3] Yeung W-CG, Rawlinson WD, Craig ME. Enterovirusinfection and type 1 diabetes mellitus: systematic reviewand meta-analysis of observational molecular studies.BMJ 2011; 342:d35; PMID:21292721; https://doi.org/10.1136/bmj.d35

[4] Richardson SJ, Willcox A, Bone AJ, Foulis AK, MorganNG. The prevalence of enteroviral capsid protein vp1immunostaining in pancreatic islets in human type 1 dia-betes. Diabetologia. 2009; 52:1143-51; PMID:19266182;https://doi.org/10.1007/s00125-009-1276-0

[5] Hyoty. Viruses in type 1 diabetes. Pediatr Diabetes 2016;22:56-64; https://doi.org/10.1111/pedi.12370

[6] Bergamin CS, Dib SA. Enterovirus and type 1 diabetes:What is the matter? World J Diabetes. 2015; 25:828-39;https://doi.org/10.4239/wjd.v6.i6.828

[7] Ja€ıdane H, San�e F, Gharbi J, Aouni M, Romond MB,Hober D. Coxsackievirus B4 and type 1 diabetes patho-genesis: contribution of animal models. Diabetes MetabRes Rev 2009; 25:591-603; https://doi.org/10.1002/dmrr.995

[8] Hober D, Sauter P. Pathogenesis of type 1 diabetes melli-tus: interplay between enterovirus and host. Nat RevEndocrinol 2010; 6:279-89; PMID:20351698; https://doi.org/10.1038/nrendo.2010.27

[9] Chehadeh W, Bouzidi A, Alm G, Wattr�e P, Hober D.Human antibodies isolated from plasma by affinity chro-matography increase the coxsackievirus B4-induced syn-thesis of interferon-alpha by human peripheral bloodmononuclear cells in vitro. J Gen Virol 2001; 82:1899-907; PMID:11457996; https://doi.org/10.1099/0022-1317-82-8-1899

[10] Hober D, Chehadeh W, Weill J, Hober C, Vantyghem M-C, Gronnier P, Wattr�e P. Circulating and cell-boundantibodies increase coxsackievirus B4-induced produc-tion of IFN-alpha by peripheral blood mononuclear cellsfrom patients with type 1 diabetes. J Gen Virol. 2002;83:2169-76; PMID:12185270; https://doi.org/10.1099/0022-1317-83-9-2169

[11] Hober D, Chehadeh W, Bouzidi A, Wattr�e P. Anti-body-dependent enhancement of coxsackievirus B4infectivity of human peripheral blood mononuclearcells results in increased interferon-alpha synthesis. JInfect Dis 2001; 184:1098-108; PMID:11598831;https://doi.org/10.1086/323801

[12] Sauter P, Hober D. Mechanisms and results of the anti-body-dependent enhancement of viral infections and rolein the pathogenesis of coxsackievirus B-induced diseases.Microbes Infect 2009; 11:443-51; PMID:19399964;https://doi.org/10.1016/j.micinf.2009.01.005

[13] Alidjinou EK, San�e F, Engelmann I, Hober D. Serum-dependent enhancement of coxsackievirus B4-inducedproduction of IFNa, IL-6 and TNFa by peripheralblood mononuclear cells. J Mol Biol. 2013; 425:5020-31; PMID:24120940; https://doi.org/10.1016/j.jmb.2013.10.008

[14] Hober D, Sane F, Ja€ıdane H, Riedweg K, Goffard A,Desailloud R. Immunology in the clinic review series;focus on type 1 diabetes and viruses: role of antibodiesenhancing the infection with Coxsackievirus-B in thepathogenesis of type 1 diabetes. Clin Exp Immunol. 2012;

168:47-51; PMID:22385236; https://doi.org/10.1111/j.1365-2249.2011.04559.x

[15] Yoon JW, Austin M, Onodera T, Notkins AL. Isola-tion of a virus from the pancreas of a child with dia-betic ketoacidosis. N Engl J Med. 1979; 300:1173-9;PMID:219345; https://doi.org/10.1056/NEJM197905243002102

[16] Elmastour F, Jaidane H, Aguech-Oueslati L, BenkahlaMA, Aouni M, Gharbi J, Sane F, Hober D. Immunoglob-ulin G-dependent enhancement of the infection withCoxsackievirus B4 in a murine system. Virulence. 2016;30:1-9

[17] Raschperger E, Thyberg J, Pettersson S, Philipson L, FuxeJ, Pettersson RF. The coxsackie- and adenovirus receptor(CAR) is an in vivo marker for epithelial tight junctions,with a potential role in regulating permeability andtissue homeostasis. Exp Cell Res. 2006; 312:1566-80;PMID:16542650; https://doi.org/10.1016/j.yexcr.2006.01.025

[18] Fechner H, Noutsias M, Tschoepe C, Hinze K, WangX, Escher F, Pauschinger M, Dekkers D, Vetter R,Paul M, Lamers J, Schultheiss HP, Poller W. Induc-tion of coxsackievirus-adenovirus-receptor expressionduring myocardial tissue formation and remodeling:identification of a cell-to-cell contact-dependent regu-latory mechanism. Circulation. 2003; 107:876-82;PMID:12591759; https://doi.org/10.1161/01.CIR.0000050150.27478.C5

[19] Ja€ıdane H, Gharbi J, Lobert P-E, Caloone D, Lucas B,San�e F, Idziorek T, Romond M-B, Aouni M, Hober D.Infection of primary cultures of murine splenic and thy-mic cells with coxsackievirus B4. Microbiol Immunol.2008; 52:40-6; https://doi.org/10.1111/j.1348-0421.2008.00002.x

[20] Baer A, Kehn-Hall K. Viral concentration determinationthrough plaque assays: using traditional and novel over-lay systems. J Vis Exp 2014;93: e52065; https://doi.org/10.3791/52065

[21] Horwitz MS, Ilic A, Fine C, Rodriguez E, Sarvetnick N.Coxsackievirus-mediated hyperglycemia is enhanced byreinfection and this occurs independent of T cells. Virol-ogy 2003; 314:510-20; PMID:14554080; https://doi.org/10.1016/S0042-6822(03)00462-8

[22] See DM, Tilles JG. Pathogenesis of virus-induced diabetesin mice. J Infect Dis 1995; 171:1131-8; https://doi.org/10.1093/infdis/171.5.1131

[23] Beck MA, Chapman NM, McManus BM, Mullican JC,Tracy S. Secondary enterovirus infection in the murinemodel of myocarditis. Am J Pathol 1990; 156:669-81

[24] Yu JZ, Wilson JE, Wood SM, Kandolf R, Klingel K, YangD, McManus BM. Secondary heterotypic versus homo-typic infection by Coxsackie B group viruses: impact onearly and late histopathological lesions and virus genomeprominence. Cardiovasc Pathol. 1999; 8:93-102;PMID:10724506; https://doi.org/10.1016/S1054-8807(98)00025-8

[25] Girn J, Kavoosi M, Chantler J. Enhancement of coxsackie-virus B3 infection by antibody to a different coxsackievirusstrain. J Gen Virol. 2002; 83:351-8; PMID:11807228;https://doi.org/10.1099/0022-1317-83-2-351

[26] Chehadeh W, Lobert P-E, Sauter P, Goffard A, Lucas B,Weill J, Vantyghem M-C, Alm G, Pigny P, Hober D.

922 F. ELMASTOUR ET AL.

Viral protein VP4 is a target of human antibodiesenhancing coxsackievirus B4- and B3-induced synthe-sis of alpha interferon. J Virol 2005; 79:13882-91;PMID:16254324; https://doi.org/10.1128/JVI.79.22.13882-13891.2005

[27] Alidjinou EK, Chehadeh W, Weill J, Vantyghem MC,Stuckens C, Decoster A, Hober C, Hober D. Mono-cytes of patients with type 1 diabetes harbour Entero-virus RNA. Eur J Clin Invest 2015; 45:918-24;PMID:26108863; https://doi.org/10.1111/eci.12485

VIRULENCE 923